Being a pharmacist I’m exposed to lots of procedural changes implemented in the name of patient safety. Tall man lettering, black box warnings, pop-up warnings for allergies, drug interactions, pregnancy indicators, lactation indicators, “high risk drugs listsâ€, shiny labels to identify sound-alike-look-alike-drugs (SALAD), separation of stock for similarly named medications, bar coding, double checking, triple checking, and so on and so forth ad infinitum. As the “IT Pharmacist†I get to see all these changes up close and personal because I’m often involved in their implementation in one way or another. Do we actually have any evidence to support using all these things?
I did a cursory search of the literature and couldn’t come up with anything solid. In fact the only interesting information I came across was from the August ISMP Medication Safety Alert: “Under ideal conditions, we– meaning all human beings–fail to perform a check correctly about 5%(1,2) of the time, and we fail to detect an error during the checking process between 5%(2) and 10%(3) of the time. While under moderate stress, our failure to detect an error during an inspection or verification process increases to about 20%.(4,5)â€. Aren’t we creating additional stress when we ask people to add an additional 10 steps to the medication administration system? I propose that we are, in fact, creating additional stress.
Looking at all the safety measures we’ve put in place, I have to ask myself if we’ve gone too far. Have we created a system destined to fail? It reminds me a little of the girl that used to sit next to me in pharmacology class in pharmacy school. She we very attentive and hung on the professor’s every word. She also loved to highlight all the important items in the textbook with all sorts of brightly colored highlighters. Her text book looked a little like a mosaic; very pretty, but pretty much useless. Based on her excellent highlighting job, it was difficult for me to decide where one important section of the text ended and another began. The whole book was one giant display of color that hurt the eyes to look at.
Our safety measures feel a little like that. Everything seems to have a pop-up warning, brightly colored sticker, or special precaution attached to it. It’s hard to tell the “dangerous†drugs from the “regular†drugs anymore. Do we really need to separate inventory, use tall man lettering and SALAD labels if we’re using bar codes? Good question. In theory I would say no. Do we really need a black-box warning pop-up at the automated dispensing cabinet (ADC) in addition to a black-box warning on the medication label and medication administration record? Again, I would have to say no. After all, the pharmacist has already reviewed the order, thus granting the nurse access to the medication in the ADC. Who’s the warning really for anyway?
I remember all the changes that I was exposed to while staffing and how many times I rolled my eyes because we had “one more thing to doâ€. Now that I have to share system changes with the pharmacists that are designed for patient safety, I’ve become the guy they’re rolling their eyes at. I don’t like being that guy. Have we crossed the threshold for patient safety and become a little silly? I can’t say for sure, but I think we’re close.
1. The Institute of Petroleum. Human reliability analysis. Human factors no. 12 briefing notes. London, England; 2003.
2. Grasha A. A cognitive systems perspective on human performance in the pharmacy: implications for accuracy, effectiveness and job satisfaction. Executive Summary Report, Report No. 062100. Alexandria, VA: National Association of Chain Drug Stores; Oct. 2000.
3. Lewis M. THERP: Technique for Human Reliability Analysis. Pittsburgh, PA: University of Pittsburgh; 2002. www.pitt.edu/~cmlewis/THERP.htm
4. System Reliability Center. Technique for human error rate prediction (THERP). Rome, NY: Alion Science and Technology; 2005.
5. Gertman D, Blackman H, Marble J, et al. The SPAR-H human reliability analysis method. Prepared for The Division of Risk Analysis and Applications, Office of Nuclear Regulatory Research, US Nuclear Regulatory Commission (NRC Job Code W6355); Washington, DC; August 2005.
Jerry the toughest thing is convincing Administration that these additional technologies require additional staff to use properly. Especially in the beginning, nursing, physician and pharmacist coverages need to be increased. Most of the time Administration feels technology allows them to decrease staffing. They want patient safety AND decreases in staffing. This is obtainable after the system is seasoned and optimized.
Chad – I think you make an excellent point, but I don’t know if adding additional staff to a bad situation makes it better. I also think we’re making decisions without looking at the overall value of the impact. We need to look at ways to simplify and streamline the process in order to create a safer healthcare environment for the patient. This is especially true if we want our pharmacists out of the physical pharmacy and performing clinical functions at the bedside. I’ve worked in several facilities now, some good, some bad. The difference has been how well the system and workflow is designed, not what warning systems have been put in place. I think this is a great topic for discussion because there is no right answer. Know what I mean?
Can you imagine if driving a car was like our medication dispensing/administration system: step up to car, verify that the door has been sealed, break safety seal on car door, read warning label, scan door key, green light indicates ok to put key in door lock, red light means wrong key, have passenger verify key inserted, unlock door, document that door is unlocked, sit in car, document seat settings and adjust where appropriate, put on seat belt, have passenger verify seat belt secure, have passenger put on seat belt, have driver verify seat belt secure, document seat belt secure, pull up computer and document seat belt is secure, scan barcode on key, green light indicates ok to put key in ignition, red light means wrong key, put key in ignition, use biometirc scan to verify you are driver, start car, acknowledge pop-up warning that fuel, gas and tire pressure are ok, put foot on break, have passenger verify foot is on break, document foot is on break, place transmission in “drive”, acknowledge pop-up telling you car is in drive, document that car is in drive, acknowledge warnings that exceeding speed limit could result in ticket, document that you acknowledged the speed limit warning, acknowledge that your passenger acknowledged the speed limit warning, document that your passenger acknowledged the warning, begin driving, acknowledge pop-ups warning you of all possible hazards along your rounte, reach destination and do everything in reverse to get out of the car. By the time you reached your destination you would throw the keys down the storm drain and set your car on fire. It’s easier to walk.